What I expected was that the counter would have a higher frequency PLL (nominally) locked to the 10 MHz reference, which could be slightly out. I assumed it would do this to get a faster than expected update rate in higher digit modes. Of course, I was wrong. I noticed the higher gate times in this video. (D'oh!)
I hate to be a nasty old Greybeard (well,actually I love being one!),but,well,----Duhh!
!
And if so compare Rubidium Frequency Standard (GPSD) of cesium frequency standard?
Then we come to know the quality of these freq stnd?
Maybe men from Agilent something to help?
Nice follow-up - Cheers Dave.
Here's a request for another, if anyone with a rubidium standard is interested in helping please.
Another idea to have some fun with the counter would be to feed 1MHz as external reference and input. Does the firmware detect such a bogus reference frequency or will it keep displaying 10MHz happily? How much deviation from 10MHz is possible?
Does anyone have experience with these FE-5680As?? http://www.ebay.com/itm/FE-5680A-Rubidium-Atomic-Frequency-Standard-Oscillator-/260839921883?pt=LH_DefaultDomain_0&hash=item3cbb4558db
I can't see the auction (work block ebay) but there are plenty of resources on the net regarding the 5680A. ISTR that there's even an effort to reverse engineer the whole thing.
As to the ebay units I'm certainly happy with mine as a general workshop frequency standard. Currently checking it against a GPDSO and when I checked mid-week the difference between the two was about 3 parts in 10
11. I had a quick look last night now thet the GPDSO has had a bit longer to settle in and I couldn't see any relative phase change between the two over about 15 minutes so they are very close. Given that the GPDSO should itself be within about 3 in 10
11 averaged over 24 hours it's looking like the Rubidium is at least as good as that.
However it is a bit of a lottery - you could easily get one with hardly any life left in the bulb or which hasn't been set up correctly. It's also sometimes hard to know which of the variants of the 5680A you'll end up with.
Does anyone have experience with these FE-5680As??
This one seems not to have 10 MHz output, only 1 PPS ( 1 pulse per second ).
Another idea to have some fun with the counter would be to feed 1MHz as external reference and input. Does the firmware detect such a bogus reference frequency or will it keep displaying 10MHz happily? How much deviation from 10MHz is possible?
Depends on the counter.
The Racal Dana 199x counters have an option which adds an extra board which will accept a 1MHz, 2MHz, 5MHz or 10MHz external standard and locks a PLL at 10MHz to the external ref - in this case it would just work at whatever accuracy the 1MHz standard had.
A more naieve implementation might just do everything 10x slower and with all the readings 10x too high.
Each FE-5680 has an output 10 MHz, the difference is that not everyone can be programmed to any frequency from 1 to 20 MHz
Another idea to have some fun with the counter would be to feed 1MHz as external reference and input. Does the firmware detect such a bogus reference frequency or will it keep displaying 10MHz happily? How much deviation from 10MHz is possible?
From manual:
The Agilent 53131A has 1, 5, 10 MHz external reference capability—to match customer’s house standard (however, the Agilent 53132A’s external reference capability is 10 MHz only).
This one seems not to have 10 MHz output, only 1 PPS ( 1 pulse per second ).
Strange, I've never seen one without the 10MHz out. Countless others on ebay with it.
This one seems not to have 10 MHz output, only 1 PPS ( 1 pulse per second ).
What does that mean? The FE-5680A has a 10MHz output; not a 1 Hz output. It's also not a pulse generator; it produces a sinusoid
This one seems not to have 10 MHz output, only 1 PPS ( 1 pulse per second ).
Strange, I've never seen one without the 10MHz out. Countless others on ebay with it.
To get 10 MHz output needs programming the unit. I do not know how, never done it.
This one seems not to have 10 MHz output, only 1 PPS ( 1 pulse per second ).
What does that mean? The FE-5680A has a 10MHz output; not a 1 Hz output. It's also not a pulse generator; it produces a sinusoid
Here is data from an other FE-5680A:
Features
Extremely low silhouette: less than 1.0 inch
Stability over temperature: 5x10-11
Frequency: 1 Hz to 20 MHz Fast Warm-up: <5 min.
Stability: To 5x10-12/ to 2x10-10/year
There are MANY variants of the FE-5680, most of which are poorly documented (proprietary). The most desirable are the 1-20 MHz programmable ones. The most common are the fixed 10 MHz sinewave output. Most but not all of these have the 1PPS output, as well. Some units are only 1PPS, some output unusual frequencies for specific applications. There are also variants with or without separate SMA RF output jacks, and units that only need the +15 power, with the +5V supply developed internally.
Some people have come across units that need 24V or 48V power, as well as units with an analog C-field trim input (and no serial communications option).
The thing is, if someone finds a suitable, working FE-5680A and adds some video amps he probably will have a decent hobby lab reference, right?
The thing is, if someone finds a suitable, working FE-5680A and adds some video amps he probably will have a decent hobby lab reference, right?
I'd say so. I did my own board instead of using a commercial amp - amplified and buffered both 10MHz and 1ppm, but also integrated the power supplies onto that board. Works great.
suggestion for what video amp choose?
Dave, how accurate are the Rigol 10Mhz out compared to your Rubidium standard. ?
Dave, how accurate are the Rigol 10Mhz out compared to your Rubidium standard. ?
Without even testing it, I can tell you it's crap. Same as the internal crystal in the Agilent counter.